From the course: AWS Certified Machine Learning - Specialty (MLS-C01) Cert Prep: 3 Modeling
Unlock this course with a free trial
Join today to access over 23,100 courses taught by industry experts.
Conclusion - Amazon Web Services (AWS) Tutorial
From the course: AWS Certified Machine Learning - Specialty (MLS-C01) Cert Prep: 3 Modeling
Conclusion
- [Instructor] Okay, let's wrap up domain three here, which is modeling. What we covered was how to frame business problems as machine learning problems. We got into how to use machine learning. Also talked about supervised versus unsupervised. Remember, unsupervised is discovering hidden features. And we also got into how to select the appropriate model. So one of the ways that I like to do this is by using AutoML Systems. Then we dived into how to train machine learning models. So, talked a little bit about compute choice optimization. Also dove into cross-validation, which can help you when you're overfitting due to selection bias. Then we dove into a little bit of a newer neuro network architecture diagram and used the TensorFlow Playground. And then finally, we got into how to evaluate machine learning models, use experiment tracking, and also talk a little bit about overfitting versus underfitting. All right, see…